President Donald Trump hosted Chinese President Xi Jinping for a state dinner that featured fricassée, sea bass and crémeux, while a roster of leading AI figures gathered at the table. Among the guests were Nvidia founder Jensen Huang and his spouse, SpaceX chief Elon Musk, AMD chair Lisa Su, Apple CEO Tim Cook, OpenAI’s Sam Altman and Greg Brockman, Google co-founder Sergey Brin and Microsoft chief Satya Nadella.
A Chinese communiqué after the meeting quoted President Xi as saying that while the two nations may compete, they can also deepen cooperation in artificial intelligence. He urged continued dialogue, the sharing of perspectives on both risks and advantages, and a joint effort to prevent misuse. Xi emphasized that AI should remain under human oversight, follow a people-centered approach and serve humanity’s progress.
The same readout attributed to President Trump a call for both sides to maintain dialogue and strengthen AI cooperation. The United States has not released its own summary of the summit. Over the weekend, Treasury Secretary Scott Bessent discussed a U.S.,China AI dialogue with his Chinese counterpart He Lifeng, proposing a direct channel for incidents that could rise to a national-security level, a concept likened to the Cold-War “red phone.”
Alec Mehta, director of the Wadhwani AI Center at the Center for Strategic and International Studies, told CNBC that the meeting appears to have produced an intention to keep the conversation going, possibly through a hotline for emergencies. He noted that such a line could become a robust tool if paired with a formal incident-reporting system, but that would require genuine commitment and high-level engagement from both governments.
Export controls on advanced Nvidia chips remain a point of contention, as Washington continues to limit China’s access to the technology. Recent Chinese chip releases and AI models have prompted Beijing to argue that U.S. restrictions have limited effect. Additional friction may stem from reports that Chinese firms have remotely accessed Nvidia hardware via overseas data centers, alongside American accusations of “distillation” practices.
The dinner took place amid heightened calls for AI safety, with OpenAI and Anthropic recently unveiling cheaper models while urging stronger alignment research. Industry moves such as Qualcomm’s launch of Android-focused chips and Google’s development of the Gemini agent illustrate the competition to create trustworthy, on-device AI. Analysts suggest that continued diplomatic engagement could shape future governance frameworks for these rapidly evolving technologies.